rocker | run docker containers with overlays and convenient options | Continuous Deployment library

I wrote the docker file like this to deploy the shiny r application

I am trying to sort the neighborhoods by their populations. I used heap sorting algorithm in C#. I created an Array which keeps the population of the neighborhoods named "arr". And created an array which keeps the name of the hoods . It works good but how can I get output of sorting with name of the neighborhoods?

Not sure what I am doing wrong, but when copying a PSCustomObject in Powershell to an array element, this is copied by reference. See this:

I'm trying to run an Rscript through entrypoint but for some reason when I do this it fails to recognize the file. Here is an example.

I found vignette on using the AzureAuth package with Shiny which seems to be a modified version of this httr Sketch of shiny + oauth.

It's unclear, to me at least, how to connect the names in that guide to what I can find in my Azure Portal.

That guide asks for tenant, app, resource, and redirect.

I presume that for tenant and app I should use the IDs from here:

For resource, do I leave in the c("https://management.azure.com/.default", "openid") or do I change it to the endpoints found in the portal, ie. https://login.microsoftonline.com//oauth2/v2.0/authorize?

I changed it to just "openid".

Should the redirect variable be the URL of my app, the callback URL or something else?

ie. https://myapp.azurewebsites.net or https://myapp.azurewebsites.net/.auth/login/aad/callback

I'm going with the former. This is all in a docker container using rocker as the base.

With all this I'm getting 500 errors after I authenticate. Of course, Azure has middleware inbetween my container and the exposed internet that is redirecting to their login regardless of anything in my code. The problem is in the handoff from being authenticated to the callback. That's where things seem to break.

Edit: Turns out I was barking up the wrong tree entirely. The problem I was having was this...

Query/headers too large on Azure if auth enabled

R Shiny app on Azure App Services with Active Directory Integration

I am going to try and explain this the best way I can, but I expect that I will have to reword a few times.

So what I have is a list of diagrams that have certain specs. I have built a table with the various diagrams and their specs in Excel. I have built a GUI using PySimpleGui for the users to interact with. There are drop down boxes with the specs in them that they will use as inputs for the code to use to search for these diagrams. Here is the code for the GUI

For a daily report Rmarkdown script that I would like to encapsulate in a docker container I face a problem with a plotly plot that is created from ggplot facet_grid and then transformed using ggplotly.

When I run the script locally on R studio my plot is working as intended:

When I run the script through my docker container, it is completely greyed out:

The data is actually there, it is just not showing:

There is no differences between the scripts except for where I run it and I actually do not know where to look for solving that bug.

Here is the ggplot+plotly part inside the rmarkdown chunk:

I have a rails application that combines some R code. It works fine, and now is the time to dockerize it. I tried creating the docker based on a ruby image, which failed installing R properly, and then the other way around (R image, installing ruby using rbenv as explained here) which failed too.

Has anyone experienced with this combo?

EDIT: I have managed to create a docker using the R image, however this requires a tiring installation of many ruby dependencies. So consider this side done.

Still, why won't the other way (installing R on ruby image) work?

Ruby-based:

I want the player to be able to choose a character so I print a list of four characters.